Re: [PATCH] perf-probe: make "perf-probe -L <function>" display theabsolute path and absolute line number

From: Masami Hiramatsu
Date: Thu Jan 13 2011 - 22:22:52 EST


(2011/01/14 4:42), Franck Bui-Huu wrote:
Masami Hiramatsu <masami.hiramatsu.pt@xxxxxxxxxxx> writes:
>
>> (2011/01/13 19:20), Franck Bui-Huu wrote:
>>> From: Franck Bui-Huu <fbuihuu@xxxxxxxxx>
>>>
>>> It should be more usefull to get the full location of the function
>>
>>> (absolute line number + full path) instead of repeating the name of
>>> the function and the start line number given by the command line.
>>>
>>> So we had before:
>>>
>>> $ perf probe -L schedule | head -n3
>>> <schedule:0>
>>> 0 asmlinkage void __sched schedule(void)
>>> 1 {
>>>
>>> and now we get:
>>>
>>> $ perf probe -L schedule | head -n3
>>> </usr/src/debug/kernel-2.6.35.fc14/linux-2.6.35.x86_64/kernel/sched.c:3813>
>>> 0 asmlinkage void __sched schedule(void)
>>> 1 {
>>
>> Indeed, it could be useful for users to see where the function is...
>>
>> However, I think that should be optional, because the output lines
>> have the relative line numbers from the function, and those numbers
>> are important for users who want to probe a specific line by using
>> function relative line numbers. e.g. "schedule:10"
>>
>> And with that option, I'd suggest to show absolute line numbers on each line.
>>
>> $ perf probe -L schedule:0-1 --by-source
>> </usr/src/debug/kernel-2.6.35.fc14/linux-2.6.35.x86_64/kernel/sched.c:3813>
>> 3813 asmlinkage void __sched schedule(void)
>> 3814 {
>>
>> Or, just show source file as an additional information.
>>
>> $ perf probe -L schedule:0-1
>> <schedule@/usr/src/debug/kernel-2.6.35.fc14/linux-2.6.35.x86_64/kernel/sched.c:0>
>> 0 asmlinkage void __sched schedule(void)
>> 1 {
>>
>> I just would like to keep the consistency of the output/input format.
>
> Well, for consistency, I thought that the additional information (given
> inside angle brackets) should always be the same: a full path and an
> absolute line number which clearly identify which source file perf-probe
> is listing.

No, that is NOT an additional information. That indicates from where those
lines are started, and also gives you a hint how you can specify the
actual probe point. For example,

$ perf probe -L schedule:10
<schedule:10>
10 rq = cpu_rq(cpu);
11 rcu_note_context_switch(cpu);
12 prev = rq->curr;

this indicates the lines started from 10th line of schedule(), and
if you want to put a new event on "prev = rq->curr;" line, you
just need to say "perf probe schedule:12"

$ perf probe -L kernel/sched.c:4077
</home/mhiramat/ksrc/linux-2.6-tip/kernel/sched.c:4077>
4077 rq = cpu_rq(cpu);
4078 rcu_note_context_switch(cpu);
4079 prev = rq->curr;

And this also gives you a hint to say (just copy & paste)
"perf probe /home/mhiramat/ksrc/linux-2.6-tip/kernel/sched.c:4079"

Since perf probe also accepts FUNC@SRC:RLN, my suggestion
also keeps that rule.
---
$ perf probe -L schedule:0-1
<schedule@/usr/src/debug/kernel-2.6.35.fc14/linux-2.6.35.x86_64/kernel/sched.c:0>
0 asmlinkage void __sched schedule(void)
1 {
---

Thank you,

--
Masami HIRAMATSU
2nd Dept. Linux Technology Center
Hitachi, Ltd., Systems Development Laboratory
E-mail: masami.hiramatsu.pt@xxxxxxxxxxx>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/